home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!spool.mu.edu!yale.edu!ira.uka.de!Germany.EU.net!materna!tb
- From: tb@Materna.DE (Torsten Beyer)
- Newsgroups: comp.unix.osf.misc
- Subject: Re: Future of DCE?
- Message-ID: <tb.727603438@materna>
- Date: 21 Jan 93 08:03:58 GMT
- References: <C0vyon.619@volvo.volvo.se> <80@mhinfo.UUCP>, <tb.727360590@materna> <kfL5tYr0BwxI5janUu@transarc.com>
- Sender: root@maternaMaterna.DE
- Lines: 56
-
- Craig_Everhart@transarc.com writes:
-
- >Nobody can predict the future very well, and you're talking about
- >projects of very different levels of maturity in the same breath (Motif,
- >DME). DME is the newest project on the plate.
-
- True, but what I meant to say was: Although I personally prefer systems
- based on an open development process (whatever that may be in detail, let's
- just assume, OSF works that way :-), the problem with these types of
- products is that their fate depends not on just one vendor (as is the case
- with proprietary systems) but on a whole bunch of them. This makes
- prediction of whatever road they might follow even harder.
-
- >I imagine that different vendors will do different things with DCE, yet
- >retain compatibility. Whether it makes its way back to ``the original
- >DCE source'' is another matter, and basically solely up to the OSF. In
-
- My fear is that Version X of any OSF product is basically the root of tree
- of different products, each claiming to be OSF/X compliant but in fact the
- products move away from each other. If they move so far as to not even
- interoperable with each other I don't know, but that might be a severe
- thread.
-
- >Clearly, you accept the ``Sun next door'' as a reference platform for
-
- That to me sounds too positive. It's not that I accept the Sun next door as
- the right means of verifying protocols. But as you say, TCP/IP is so
- widespread that from a practical standpoint to veryfy X's TCP/IP port you
- can virtually use any system next door to prove interoperabiltiy. I very
- aware that this wasn't the case maybe 7 years ago. But the situation has
- changed in some way. TCP/Ip was new. There wasn't much else to build
- heterogeneous networks. Today we have an awful lot of tools and protocols
- intended to help us build and maintain these networks (wether they succeed
- is a different question altogether :-). Now here comes OSF and tells us, to
- throw away this old rubbish and start usin their stuff. I'd rather like to,
- but then how much do I pay for it. I certainly can't check X's DCE
- implementation as easy as I can with a plain and easy ONC-RPC
- implementation. I have to buy validation suites and all that. It clearly is
- a spinnoff problem. As soon as the number of available DCE implementations
- exceeds a critical mass, we'll see the same situation as with TCP/IP.
-
- >I'm sure that there will be companies like that around, but indeed the
- >stuff should just work (well, mostly).
-
- Hmm, from what you write I guess you have been in this business for quite
- some time. You should know, that every sentence containing the phrase
- "should work", simply means it doesn't :-)))))
-
-
-
- -Torsten
- --
- Torsten Beyer mail : tb@Materna.DE
- Dr. Materna GmbH vox humana : +49 231 5599 225
- Vosskuhle 37, D-4600 Dortmund fax machina : +49 231 5599 100
- "Once in a moment the SUN goes down, protect and survive" -- Runrig
-